Masterstudiengang "Drug Regulatory Affairs"

Master-Thesis

Artificial intelligence and Big Data in Pharmacovigilance. Current development, problems and perspectives. Regulatory aspects.

Olga Rassokhina (Abschlußjahr: 2022)

Summary
Language: English
Novel technologies are the main drivers of progress. Artificial intelligence (AI) is at the forefront of development in information technologies (IT). AI has infiltrated many aspects of our daily life and insensibly become a routine at work. Advantages of IT have been already implemented and became a standard in pharmacovigilance (PV): implementation of electronic submission and documents review, electronic individual case safety reporting, development of user-friendly regulatory interfaces, etc. Nonetheless, AI-based methods and tools that can be used in PV are still at the initial stage of development. Positive and negative aspects of emerging technologies have to be thoroughly considered.
The discussion process on the evaluation pros and cons has started all around the globe, and initial steps on a way to regulate AI have been taken. The existing EU regulatory framework ensure fundamental rights related to the product safety and liability, consumer rights, equal treatment in employment and occupation, personal data protection and privacy; however, doesn't fully cover all the emerging aspects triggered by the novel technologies. Therefore, the European Commission and the High-Level Expert Group in collaboration with all stakeholders provided a White Paper and presented first definitions of AI, an overview of the critical areas and risks associated with AI implementation. Consolidated opinion on multiple challenges, including ethical aspects, data security, protection of fundamental rights and the ways to overcome foreseeable bottlenecks were proposed after numerous public consultations. The European regulatory framework should build trust to AI, promote innovation, support competitiveness, ensure socially, environmentally and economically optimal outcomes and compliance with EU legislation, principles and values.
Meanwhile, pharma business has already started to use the advantages of AI. More than 80% of the leading 100 pharmaceutical companies plan to initiate a re-examination of their safety platforms using AI in the nearest future. Information management has become an integral part of the European Medicine Agency (EMA) Telematics Strategy up to 2025 and beyond. Various projects focused on collection and processing of huge data sets (Big Data (BD)) containing safety information were launched recently by EMA, World Health Organization and other international bodies. Some of these initiatives already showed their efficacy in the context of the COVID-19 pandemic.
Utilization of machine learning, deep learning, natural language processing, BD, predictive analytics in pre- and post-marketing drug safety surveillance to identify potential safety issues may improve data quality, enhance safety reporting, signal detection and benefit-risk assessment that will improve safety and efficacy of medicinal products. Data mining and BD analytics will allow taking rapid and thorough regulatory decisions based on robust analysis and efficient evidences, unified criteria for divergent sources and content.
Nevertheless, novel technologies will bring new challenges that have to be resolved: validation of BD and immense data sources, establishment of standards for AI signal detection and validation, additional investment in the infrastructure and transformation of current PV systems, adjustment of regulatory framework to a digital world. Despite significant challenges and add-on risks, implementation of AI is envisaged to be beneficial for many areas in PV.
Further development and implementation of AI in PV will help to overcome current challenges and limitations, will improve data quality and consistency in the data analysis, enhance safety reporting and signal detection, minimize risks associated with drug safety, enable change the focus in research and development area to patient-oriented outcome. Hence, it may improve quality of MPs, access to medicine all around the globe, contribute to the ultimate goal to protect public health.  
Pages: 98
Annexes: 0,  Pages: 0